Server Crashing for unknown reason across different artifact versions

Server Owners
1

Hi,

For a long time now we’ve had problems with our Linux hosted FiveM servers crashing.
Initially we figured it would be some kind of resource error, but after fixing up alot of our janky resources the problem continues.
This has been over many different artifacts and we just can’t seem to find the problem.
I’ve tried opening the dump file (in visual studio) but I’ve got little idea on what its actually displaying.

Current artifacts version: 2907 Linux (But we’ve tried recommended ones in the past as well).

Download dump file:
https://drive.google.com/file/d/1P3gCORqFR2A2UUVJMd9ISn5Q8nuoKqSA/view?usp=sharing

I would love some more insight into this issue as it has been causing our server to crash after about 2 days of uptime.
We don’t run any heavy RP frameworks as the server is meant for Drifting vehicles around tracks and Los Santos map.

If there is any more information required please let me know and I will try to respond in a timely manner.

Thank you for your time :heart:

2

I’ve had similar problems. Can you explain how your server is setup on your linux machine and also how do you go by starting the server?

3

I don’t think Linux dump files contain any relevant info at all. :confused: Do you have the reported crash ID from console output?

4

53643d45-1c13-42af-91ac-132a89e4e4dd

Is the “Crash report ID” it gave me.

Image with some more lines from the log output:

image
image989×206 8.51 KB

We normally don’t get any kind of Sync or Network thread hitches either, so I’m guessing it has something to do with it. From time to time we get a server thread hitch warning, but that’s quite rare.

5

Our Dedicated machine:
OS: Xubuntu Desktop 16.04 “Xenial Xerus”
CPU: Intel i7-7700K
RAM: 32GB

The way we start our servers is through a Bash script which checks if the process exists, if not it starts it.

It will run a command using “xfce4-terminal” to create a new terminal window which then runs the normal starting command “PATH_TO_SERVER/run.sh +exec server.cfg”.

We are running multiple servers from a single folder, but they are restarted in a way where only 1 server is generating cache before others restart when we do updates for example.

If you need any more info let me know. I really would like to fix this problem if possible at all.

6

Seems to be a crash in libcurl.so.4 when something is called from OnTimeout in http-client. Not too common a crash, either, nor one that has much info of note logged.

Might be something like this one: https://github.com/curl/curl/issues/4120

… which’d mean we’d somehow have to update the system alpine libcurl.

… which should be done in server build 2943.

7

Oh okay, damn I didn’t expect that.

I will tryout the 2943 build and report back.
It might take a few days to see if it still crashes.

However, thank you for your time so far! It’s greatly appreciated :heart:

8

Okay so we updated to 2943 a few days ago.

The server still crashes, and it has created a lot of Log entries (like 10 000s of entries).
All of them saying:

error on connection: too many open files
Listening on socket failed - libuv error EMFILE.

There doesn’t seem to be anything happening right before it with hitches or something like that.
Also there was no crash code from what I can see.

Should I update to 2949?

9

Increase numfile from your distro’s default.

10

I’ve increased it to 3 000 000 I think that should be enough right?

Let’s see what happens, thank you again for your help.

11

Update on this:
This error(below) doesn’t seem to be a fault in the server, but someone DDOSing our servers! :partying_face:

error on connection: too many open files
Listening on socket failed - libuv error EMFILE.

So far we haven’t been able to keep the servers online long enough to have them crash in their usual time-frame.